SI4706-C31 Datasheet, PDF (18/36 Pages) Silicon Laboratories – HIGH-PERFORMANCE FM RDS/RBDS RECEIVER

Si4706-C31
The Si4706 performs much of the FM demodulation
digitally to achieve high fidelity, optimal performance
versus power consumption, and flexibility of design. The
on-board DSP provides unmatched pilot rejection,
selectivity, and optimum sound quality. The integrated
micro-controller offers both the manufacturer and the
end-user unmatched programmability and flexibility in
the listening experience.
4.2. FM Receiver
The Si4706 FM receiver is based on the proven
Si4700/01/02/03 FM radio receiver. The part leverages
Silicon Laboratories' proven and patented FM broadcast
radio receiver digital architecture, delivering excellent RF
performance and interference rejection. The proven
digital techniques provide good sensitivity in weak signal
environments while providing superb selectivity and
inter-modulation immunity in strong signal environments.
The part supports the worldwide FM broadcast band (76
to 108 MHz) with channel spacings of 50–200 kHz. The
Low-IF architecture utilizes a single converter stage and
digitizes the signal using a high-resolution analog-to-
digital converter. The audio output can be directed
either to an external headphone amplifier via analog
in/out or to other system ICs through digital audio
interface (I²S).
4.3. Stereo Audio Processing
The output of the FM demodulator is a stereo
multiplexed (MPX) signal. The MPX standard was
developed in 1961, and is used worldwide. Today's
MPX signal format consists of left + right (L+R) audio,
left – right (L–R) audio, a 19 kHz pilot tone, and
RDS/RBDS data as shown in Figure 10 below.
4.3.1. Stereo Decoder
The Si4706's integrated stereo decoder automatically
decodes the MPX signal using DSP techniques. The 0
to 15 kHz (L+R) signal is the mono output of the FM
tuner. Stereo is generated from the (L+R), (L–R), and a
19 kHz pilot tone. The pilot tone is used as a reference
to recover the (L–R) signal. Output left and right
channels are obtained by adding and subtracting the
(L+R) and (L–R) signals respectively.
4.3.2. Stereo-Mono Blending
Adaptive noise suppression is employed to gradually
combine the stereo left and right audio channels to a
mono (L+R) audio signal as the signal quality degrades
to maintain optimum sound fidelity under varying
reception conditions. Three metrics, received signal
strength indicator (RSSI), signal-to-noise ratio (SNR),
and multi-path interference, are monitored
simultaneously in forcing a blend from stereo to mono.
The metric which reflects the minimum signal quality
takes precedence and the signal is blended
appropriately.
All three metrics have programmable stereo/mono
thresholds and attack/release rates detailed in AN344
Revision 0.3: Si4706/07/4x Programming Guide (NDA)
and greater. If a metric falls below its mono threshold,
the signal is blended from stereo to full mono. If all
metrics are above their respective stereo thresholds,
then no action is taken to blend the signal. If a metric
falls between its mono and stereo thresholds, then the
signal is blended to the level proportional to the metric’s
value between its mono and stereo thresholds, with an
associated attack and release rate. Stereo/mono status
can be monitored with the FM_RSQ_STATUS
command.
4.4. Received Signal Qualifiers
Mono Audio
Left + Right
Stereo
Pilot
Stereo Audio
Left - Right
RDS/
RBDS
0
15 19 23
38
53 57
Frequency (kHz)
Figure 10. MPX Signal Spectrum
A tuned signal's quality can vary with the environmental
conditions, time of day, and position of the antenna
among many other factors. To adequately manage the
audio output and avoid unpleasant audible effects to the
end-user, the Si4706-C31 monitors and provides
indicators of the signal quality, allowing the host
processor to perform additional processing if required
by the customer. The Si4706-C31 monitors and reports
a set of standard industry signal quality metrics
including RSSI, SNR, and multi-path interference on FM
signals. As with other Si4706-C31 features, how these
variables are used to improve audio performance can
be left to the Silicon Labs default on-chip algorithms
(recommended), or they can be customized to modify
the performance of the part.
18
Rev. 1.0